[90] Suidas, sub Anast. Mordtmann (op. cit., p. 33), thinks the ruins existing at Tekfur Serai may represent the original Palace of Blachernae, the basement, at least. It is commonly called the palace of Constantine, etc., but Van Millingen proves it to be a late erection.
[91] Zonaras, xiii, 24; Codin., p. 95, etc.
[92] Const. Porph., De Cer. Aul. Byz., ii, 12. Still frequented (Paspates, op. cit., p. 390, etc.).
[93] To “a man’s height” (Paspates).
[94] Paspates has all the credit of solving the problem of this moat (op. cit., p. 7, etc.). It has been maintained that it was a dry moat, owing to the physical impossibility of the sea flowing into it. The words of Chrysoloras (Migne, Ser. Grk., vol. 156, etc.) are alone sufficient to dispose of this error.
[95] This space seems to have been called the παρατείχιον; Const. Porph., loc. cit.; or rather, perhaps, the πρωτείχισμα; see the Anon., Στρατηγική (Koechly, etc.), 12 (c. 550). Paspates calls it the προτείχιον, “because,” says he, “I have found no name for it in the Byzantine historians.”
[96] Ducas, 39, etc.; Paspates, op. cit., p. 6. It is, however, the usual word for the walls of a city. Μεσοτείχιον and σταύρωμα are more definite; Critobulos, i, 60. Paspates states that the ground here has been raised six feet above its ancient level.
[97] Déthier, Nouv. recherch. à CP., 1867, p. 20; cf. Vegetius, iv, 1, 2, 3, etc. These walls have much similarity to the agger of Servius Tullius, but in the latter case the great wall forms the inner boundary of the trench and the lesser wall, retaining the excavated earth, was about fifty feet behind in the city. See Middleton’s Ancient Rome, etc.
[98] Paspates, op. cit., p. 17.
[99] Ibid., Grosvenor, op. cit., p. 584.